Precision agriculture ’09 659 Modelling competition for below-ground resources and light within a winter pea (Pisum sativum L.) – wheat (Triticum aestivum L.) intercrop (Azodyn-InterCrop): towards a decision making oriented-tool P. Malagoli 1 , C. Naudin 1 , G. Goulevant 1 , M. Sester 1 , G. Corre-Hellou 1 , M.-H. Jeuffroy 2 1 Laboratoire d’Ecophysiologie Végétale et Agroécologie, Ecole Supérieure d’Agriculture, 55 rue Rabelais, BP 30748, 49007 Angers Cedex 01, France; p.malagoli@groupe-esa.com 2 INRA UMR 211 Agronomie INRA AgroParisTech, BP 01, 78850 Thiverval-Grignon, France Abstract Grain legume-cereal intercrops allow a gain of productivity grown along the growth cycle on the same piece of land under low input (of which nitrogen (N) fertilizers) levels. This is partly due to a better use of soil nitrogen (larger available soil N per plant for cereal N uptake and an increased contribution of N ixation for pea nutrition) under combinations (species and crop management systems) fully optimized within a given soil and climate environment. Modeling is a powerful tool to explore a wide range of combinations. It can be further used as a decision making oriented-tool provided below-ground resources and light sharing is satisfactorily simulated. Our work aimed at designing a new dynamic intercrop growth model (Azodyn Inter-Crop (IC)) based upon Azodyn for wheat and Aisol for pea. Nitrogen and water partitioning between species is irstly driven by nitrogen and water demand of each species. When intercrop demand is larger than soil supply then water and N acquisition is limited by root exploration, soil nutrient supply and N taken up by the companion species as it concurrently depletes available below-ground resources. The ‘functional’ root layer concept allows to account for advantage towards species with a faster root penetration rate. Leaf area expansion is driven by daily satisfaction of N demand, itself computed through an adapted version of N dilution curve to intercrop growth. Light sharing depends on leaf area index (LAI) growth and leaf properties (relectance, leaf angle) of each species. Model outputs show Azodyn-IC can satisfactorily simulate N taken up, LAI, light interception eficiency and crop growth of each sole- and intercropped species along the growth cycle leading to realistic yields for the applied N fertilizer rates. It also emphasizes competition for light and below-ground resources within intercrops is tightly and dynamically linked within intercrop. Keywords: cereals, legumes, mixture, growth, simulation Introduction Intercropping consists of growing simultaneously two or more species on the same piece of land. This cropping system results in a more eficient use of available resources (such as soil water and nitrogen) by intercrops than sole crops (see Malézieux et al., 2008 for a review). It has been extensively reported in spring pea-barley (Jensen, 1996a; Corre-Hellou et al., 2006) and spring pea-wheat (Ghaley et al., 2005) intercrops. This advantage can be mainly explained by the complementary use of N (N derived from soil N uptake or from N 2 ixation) in intercrops. Cereals have an earlier date of emergence, a higher demand for nitrogen and a faster root penetration than pea (Corre-Hellou and Crozat, 2005). Accordingly they have a competitive advantage to take below-ground resources (water and nitrogen) up early in the growth cycle. This advantage is strengthened by the fact (1) cereal plant density is half of sole-cropped cereals into replacement intercropping system and (2) grain legumes massively acquire N through atmospheric N ixation (47% of total N uptake in sole crop up to 91% for pea grown with spring barley; Corre-Hellou et al., 2006). Competition for N